00:00:00 --- log: started forth/02.01.16
00:16:00 * aaronl is away: I'm busy
00:34:47 * aaronl is away: sleep
09:29:43 --- join: PoppaVic (pfv@d169.as0.gylr.mi.voyager.net) joined #forth
09:29:59 <PoppaVic> Anyone had any luck with the latest pfe?
10:02:33 --- part: PoppaVic left #forth
10:44:22 --- quit: clog (^C)
10:44:22 --- log: stopped forth/02.01.16
10:45:05 --- log: started forth/02.01.16
10:45:05 --- join: clog (nef@bespin.org) joined #forth
10:45:05 --- topic: 'do drop >in'
10:45:05 --- topic: set by I440r on [Tue Jan 15 08:40:12 2002]
10:45:05 --- names: list (clog oxygene aaronl @Speuler)
13:55:48 --- quit: Speuler (Read error: 113 (No route to host))
14:01:00 --- join: Speuler (~l@c38038.upc-c.chello.nl) joined #forth
14:10:55 --- quit: aaronl ("hardware installation")
14:25:27 --- join: aaronl (aaronl@vitelus.com) joined #forth
14:36:33 --- quit: aaronl ("new kernel image")
14:38:45 --- join: aaronl (aaronl@vitelus.com) joined #forth
15:21:36 --- quit: aaronl ("The name's X. Bitch X.")
15:21:48 --- join: aaronl (aaronl@vitelus.com) joined #forth
18:57:28 --- join: futhin (thin@h24-66-212-245.cg.shawcable.net) joined #forth
18:57:31 <futhin> hi all
18:57:43 <futhin> hi speuler
18:57:45 <futhin> hi oxygene
18:57:51 <Speuler> hi futhin
18:58:00 <futhin> how is it going speuler? :)
18:58:07 <Speuler> not too bad.
18:58:10 <Speuler> just returned
18:58:19 <Speuler> setting up new machine as firewall
18:58:29 <Speuler> needs an upgrade
18:58:58 <Speuler> different distro
18:58:58 <futhin> returned?
18:59:07 <Speuler> friend needs same
18:59:17 <Speuler> so we figured a few things out together
18:59:26 <Speuler> actually have my machine there
18:59:40 <Speuler> generic install to be copied to 2nd machine
19:00:04 <Speuler> smoothly going as far
19:00:23 <futhin> i downloaded a pygmy tutorial that is combined with pygmy, it has been helpful in getting me back into forth :)
19:00:53 <Speuler> tutorial is forth program ?
19:01:59 <futhin> the tutorial is coded into pygmy..   it shows you how to use the editor, types in code for you, shows result, etc..  and displays the textual part of the tutorial
19:02:16 <Speuler> nice
19:02:33 <Speuler> you can try out the examples while the tutorial is visible on screen ?
19:03:25 <futhin> nope, not quite
19:04:01 <futhin> it's not the best, and there's some grammatical errors, but it's decent enough to bring me back into forth, because i've been so lazy about coding in it
19:04:08 <futhin> my z key is dying on me!
19:06:32 <Speuler> what are your plans with forth ?
19:06:49 <futhin> i've got a lot of ideas :P
19:07:25 <futhin> editor, debugger, interactive forth collaborative online environment, forthmud, forthos, etc :)
19:07:49 <futhin> i'm sort of working on the forthmud, but it's been dead for 2 months
19:07:50 <Speuler> well,  that's one :)
19:08:26 <Speuler> i remember the network socket interface you were looking for
19:09:37 <futhin> yeah
19:09:44 <Speuler> are you linked anyway to the comp.lang.forth article, concerning writing linux in forth ?
19:09:46 <futhin> i've been avoiding figuring out the socket interfacing
19:09:53 <futhin> eh? nope
19:10:39 <futhin> i'm more interested in a pure forth os, properly designed from the ground up, targetted to the desktop market
19:10:50 <Speuler> valdocs II
19:11:36 <futhin> eh?
19:11:39 <futhin> valdocs II ?
19:12:06 <Speuler> i'm just looking for a suitable pointer. valdocs (epson) was a forth os
19:12:09 <Speuler> never quite finished
19:12:21 <Speuler> eventually epson gave up
19:12:55 <Speuler> too slow. machines weren't suited for non-asm os by then
19:13:03 <futhin> lol
19:13:18 <Speuler> http://legalminds.lp.findlaw.com/list/cyberia-l/msg16267.html
19:13:28 <futhin> how do i fix my keyboard? my z key is dying, and my alt key is already dead :(
19:13:51 <Speuler> use the 2
19:13:55 <Speuler> such as in  la2y
19:14:56 <futhin> lol
19:15:11 <futhin> the url isn't much of a reference to vadocs :(
19:15:33 <Speuler> one of the few refs i found as far
19:16:07 <futhin> yeah
19:16:26 <futhin> i need to give cpr to my keyboard, or buy a new one, and buy a keyboard cover!!
19:17:05 <futhin> i don't know why keyboards don't come with a keyboard cover?  maybe it is a conspiracy to let keyboards get ruined by spills
19:18:41 <Speuler> no more refs to valdocs found. looks dead
19:18:48 <futhin> yeah
19:18:58 <futhin> there are a few other forth oses out there i think
19:19:02 <Speuler> wasn't a success
19:19:16 <Speuler> forth is its own os
19:19:35 <futhin> yeah i know
19:19:49 <futhin> but i mean a forth that is developed up to the point that users can use it
19:20:11 <Speuler> 2 list
19:20:40 <Speuler> 3 block b/blk erase update
19:21:07 <Speuler> 2 10 index
19:21:27 <Speuler> what else does a user need :)
19:21:35 <futhin> heh
19:22:07 <futhin> games! full 3dimensional first person shooters! and strategy games! ones that professional military men will be awed by!
19:22:23 --- join: aaronl_ (aaronl@vitelus.com) joined #forth
19:22:26 <Speuler> but, that's not part of an os. thats application programs
19:22:41 --- quit: aaronl (Read error: 113 (No route to host))
19:22:43 <Speuler> why you write a dos if you want a game ?
19:23:23 <futhin> the os needs to come with applications and utilities in order to be attractive to users
19:23:38 <Speuler> i hardly ever play
19:23:47 <Speuler> computer games
19:23:50 <futhin> yeah, same here
19:23:55 <futhin> but that was just an example
19:24:00 <Speuler> not a good sales point 
19:24:06 <Speuler> sell ?
19:24:09 <futhin> the os needs a gui
19:24:11 <Speuler> give away ?
19:24:12 <futhin> networking
19:24:14 <futhin> internet
19:24:29 <futhin> i dunno
19:25:10 <futhin> non-crippled shareware, opensource.  register for 10 bucks to get some extra stuff or something
19:25:41 <Speuler> if an os needs a game to become attractive, the os itself failed to be attractive
19:26:01 <futhin> heh
19:26:03 <Speuler> gui is optional stuff
19:26:34 <futhin> it's a desktop os, aimed at normal people who don't know anything about computers.. 
19:26:43 <futhin> i want to save their souls from microsoft
19:27:02 <Speuler> excluding non-crippled shareware requires some license to restrict people from writing crippled shareware for the os 
19:27:20 <Speuler> or disallow writing for the os altogether 
19:27:29 <futhin> eh?
19:27:36 <futhin> disallow writing for the os altogether?
19:27:58 <Speuler> how to keep people from offering crippled shareware ?
19:29:11 <Speuler> "any program offered to run under os xyz must be distributed free of charge" ?
19:29:30 <futhin> i don't care about them, i don't need to care about them.  the os will be so elegant and simple, that normal users will be able to code themselves, to modify the os at will.   crippled shareware won't survive
19:30:04 <futhin> but why should i prevent other ppl from coding crippled shareware?  i want the majority of people to use the os..
19:30:13 <Speuler> how can the os influence decision of program writers, not to cripple their sw ?
19:30:47 <futhin> because anybody will be able to code easily
19:31:08 <futhin> if some guy comes out with some shareware, his idea will be easily replicated by other ppl
19:31:18 <Speuler> in forth ? everybody ?
19:31:40 <Speuler> what a horrible thought
19:31:44 <Speuler> hehe
19:32:09 <futhin> yup.. i'll create several layers of abstraction.  the highest layer will be extremely high-level and extremely simple to use, it won't be fast, but my grandma will be able to use it. 
19:32:31 <Speuler> one might even easily write a c interpreter, to run the obsolete c software base.
19:32:46 <futhin> naw
19:32:49 <futhin> why?
19:32:56 <futhin> screw c
19:33:02 <Speuler> some people are la2y
19:33:22 <futhin> yeah well, the la2y ppl can code the c interpreter themselves! :P
19:33:32 <Speuler> they think "why should i rewrite this when i got it already"
19:33:58 <futhin> because it'll be faster, more consistent, and will be a breeze to code it in the forth os
19:34:03 <Speuler> then they offer it as crippled shareware
19:34:18 <futhin> lol
19:34:36 <Speuler> forth os polluted
19:35:25 <futhin> i was thinking of creating a utility that is distributed with the forth os, that can be used to check the purity and efficiency of other programs.  that way the users can check a program to see if it is efficient and if its not they can get rid of it ..
19:36:22 <Speuler> now one writes an asm program
19:36:32 <Speuler> for the forth os
19:36:51 <futhin> yeah i dunno about asm
19:37:11 <futhin> maybe write some ugly disasm that converts the code into forth.. lol
19:41:16 <Speuler> what's a forth os good for if you can run a forth interpreter under an existing os ?
19:41:36 <Speuler> (and do all you'd like the os to do through the interpreter ?)
19:42:00 <Speuler> save memory ?
19:42:03 <futhin> faster, one layer between user and computer
19:42:08 <Speuler> yes ram prices went up
19:42:09 <futhin> consistency
19:42:37 <futhin> enable users to access any part of the os with ease and modify it, add to it
19:43:24 <Speuler> the os contains its own source code.
19:45:08 <futhin> speed is one of my priorities.  i have had too many bad experiences with windows and linux
19:45:30 <Speuler> valdocs failed because of lack of speed
19:45:45 <Speuler> cpus weren't as quick then as they are now
19:47:03 <futhin> yeah
19:48:47 <Speuler> i find linux pretty quick, without gui
19:49:05 <Speuler> and enough ram provided
19:49:28 <Speuler> hard disk is slow
19:49:38 <futhin> i've always had the slower computer, i've always been a few years behind on the upgrade curve.  so when windows has trouble on my computer, with being responsive, having multiple programs open, i get pissed off.  especially when there are programs that do very little and are _noticeably_ bloated.
19:49:59 <Speuler> what cpu do you run now ?
19:50:13 <futhin> 133 mhz, 32 megs of ram.  it used to be 16 megs.
19:50:15 <futhin> pentium
19:50:44 <Speuler> new firewall is a 486 dx2-66
19:50:49 <Speuler> 20 mb ram ...
19:51:03 <Speuler> current is 133 mhz p1, 64 mb ram
19:51:28 <Speuler> i have x running on a 386-40, with 8 mb ram :)
19:51:43 <futhin> a gui is necessary for me because i require a graphical web browser
19:51:58 <futhin> i tried to get x working on 16 megs, but it always complained :(
19:52:14 <Speuler> 8 mb ram, without swap !
19:52:17 <futhin> lol
19:52:36 <Speuler> but, its a remote session
19:52:51 <Speuler> just linux, and xserver. rest indirect
19:53:22 <Speuler> desktop, window manager, x apps and the like consume mem
19:53:40 <Speuler> the 386/8m would be far too small for those
19:54:05 <Speuler> but, just some screen updates it can do perfectly well
19:54:51 <Speuler> it happens that people find that they need a quicker machine at home, seeing the "speed" of that 386 ...
19:55:04 <futhin> there's this guy saying that it is impossible to add Garbage Collection to Forth..  what do you think?
19:55:13 <Speuler> and when i tell them that that machine is a 386 with 8 mb, they are stunned
19:55:37 <Speuler> GC: yes and no
19:56:08 <futhin> hmm?
19:56:17 <Speuler> GC without forth extensions to make use from GC makes little sense
19:56:37 <Speuler> AND would be difficult to implement
19:57:17 <futhin> eh?
19:57:21 <futhin> that doesn't make any sense to me
19:57:57 <Speuler> but, say you put a layer on top of forth, giving you a programming language/environment where dynamically allocated memory is used more than it is in forth
19:59:01 <Speuler> that system could be able to profit from a mem man system, supporting gc
19:59:18 <futhin> jeez, the guy is saying that forth is a low-level language, that GC is the difference between a high-level language and a low-level language.   what if GC is completely arbitrary to forth ??
20:00:19 <Speuler> no, i say that standard forth has little to take advantage of gc
20:00:39 <futhin> you said "GC without forth extensions to make use from GC.. "   we aren't talking about an external GC program.  we're talking about garbage collection inside forth..
20:00:58 <futhin> is forth low-level or high-level?  can it be high-level ?
20:01:11 <Speuler> it is as low or high the programmer want it
20:01:23 <Speuler> it can be very high level
20:01:29 <Speuler> by extending it
20:05:15 <futhin> yeah, but if it is extended, is it a "high-level" language?
20:05:19 <Speuler> a, say, standard variable in forth is not a sufficiently dynamic contruct to disappeat if it is not used any longer.
20:05:23 <futhin> what defines a high-level language? .. 
20:05:46 <Speuler> nor are forth words made to suddenly change, so they don't refer to a certain data item anymore
20:06:30 <Speuler> if your extensions introduce that kind of dynamicism, gc could be helpful;
20:07:16 <Speuler> higher-level = higher level of abstraction of concepts introduced by program.
20:07:38 <futhin> hmm
20:07:55 <futhin> i've gotta go now heh
20:08:12 <Speuler> ok.
20:08:18 <Speuler> getting l8 for me too
20:08:31 <Speuler> 5 a.m
20:09:16 <Speuler> g'd night
20:10:27 <futhin> ok
20:10:28 <futhin> good night
20:10:29 <futhin> lol
20:10:30 <futhin> 5am
20:10:34 <futhin> that's pretty late :P
20:10:57 <futhin> laters
20:11:01 --- quit: futhin ("bye, g2g")
21:30:31 --- nick: aaronl_ -> aaronl
22:45:50 --- join: kholmes (~kholmes@client543.sedona.net) joined #forth
23:17:34 --- join: Xuz (aemerson@bgp01079860bgs.wanarb01.mi.comcast.net) joined #forth
23:17:40 --- mode: ChanServ set mode: +o Xuz
23:18:38 * Xuz waves
23:19:28 * kholmes waves back.
23:20:16 <Xuz> How're you?
23:21:02 <kholmes> okay
23:21:05 <kholmes> yourself?
23:21:11 <Xuz> Oh, fine.
23:21:28 <kholmes> so...you code forth?
23:22:20 <Xuz> Yep
23:22:46 <kholmes> I don't...but I'm curious about it
23:22:53 <kholmes> what do you use it for?
23:24:56 <Xuz> Anything I can, at the moment.  Unfortuantely with the Forth I have access to (GForth) that's mostly pure computation.
23:25:32 <kholmes> what you mean by pure computation? you mean numerical calculations?
23:28:35 <Xuz> Well. . . pretty much anything that doesn't require me linking in libraries.
23:29:16 <Xuz> Which mostly just makes I/O (curses, X, sound, networking, etc.) a pain.
23:29:41 * kholmes installs gforth
23:29:52 --- join: I440r____ (~mark4@1Cust116.tnt2.bloomington.in.da.uu.net) joined #forth
23:29:59 <I440r____> hi ppl@!
23:30:04 --- nick: I440r____ -> I440r
23:30:17 <kholmes> hi
23:30:43 --- mode: ChanServ set mode: +oo aaronl clog
23:30:43 --- mode: ChanServ set mode: +oo kholmes oxygene
23:30:43 --- mode: ChanServ set mode: +oo Speuler I440r
23:30:58 <Xuz> Great for applications that you want to interact with mostly textually, be able to program and whatnot, and don't require anything fancy in the way of I/O or fiddling with big, complicated structures that people have already written libraries for, though.
23:31:53 <Xuz> GForth's library interface is, unfortunately, rather unpleasant to use, undocumented, and only lets you call library functions with six arguments or less.  And there's no good way to pass a word to a function requiring a function pointer that I've found.
23:32:43 <kholmes> hmm
23:32:46 <Xuz> Er, an XT, pardon me.
23:32:56 <Xuz> But, for alot of things you don't need that.
23:32:58 <kholmes> forth doesn't have module system does it?
23:33:08 <Xuz> It can if you want :)
23:33:16 <Xuz> You can include files.
23:33:29 <kholmes> okay
23:33:38 <Xuz> And you can build and use as much of a module system as you like and make them work with it.
23:33:39 <I440r> i dont like using other ppls libraries. and i rarely write any of my own. if i need a function i write it 
23:33:47 <I440r> i dont write jack of all trades code
23:33:56 <I440r> and i care nothing for portabllity
23:34:24 * Xuz does care for portability, however, which is why he got (most) of a curses interface up and running instead of outputting terminal control codes.
23:34:51 <Xuz> Especailly since even if I'm only ever going to run a program on one type of machine, I don't know what kind of machine the user will be connecting from.
23:34:53 <kholmes> well...I guess postfix syntax is interesting to me...it seems more natural
23:35:15 <kholmes> Xuz: cool
23:35:52 <I440r> hehe
23:36:07 <Xuz> Oh, it's wonderful stuff.
23:36:53 <Xuz> I just wish you weren't required to hunt down undocumented stuff and fiddle about with libc to open a network connection :)
23:37:02 <Xuz> As for anything with callbacks. . . just don't bother.
23:37:26 <I440r> if i ever get back into isforth thats one of my prime directives
23:37:33 <I440r> NO undocumented functions
23:38:27 <Xuz> kholmes: What kind of things have you bee reading on Forth so far?
23:38:57 <kholmes> well...its been a while ago since I looked into forth
23:39:16 <Xuz> It's lovely, despite my rants about one particular implementation :)
23:39:22 <kholmes> but there has been some discussion on the Tunes list
23:39:32 <kholmes> they say its low-level
23:39:51 <I440r> its both low level AND high level
23:39:55 <I440r> i would cal it broad spectrum
23:40:08 <I440r> it can be as high level as ada and as low level as assembler
23:40:22 <kholmes> hmm...as high level as other high level languages?
23:40:44 <Xuz> Yes and no.
23:40:47 <I440r> yes
23:40:59 <I440r> you can code as high level as you want
23:41:13 <Xuz> It has a very high power of abstraction.  You can get sort-of closures, sort-of first-class-anything-you-like, even kind-of-continuations. . . 
23:41:36 <Xuz> But you still have to deal with things like memory management on your own.  And there's not much in the way of dynamic checks, which is fine with me.
23:41:39 <kholmes> what do you mean by kind-of/sort-of
23:41:50 <kholmes> oh
23:42:31 <I440r> there shouldnt be any checks
23:42:35 <Xuz> As in, you can get some of the exact same benefits as first-class functions with somewhat odd syntax.  You can get closures out of create/DOES>
23:42:42 <I440r> i dont even think there should be any stack checks
23:42:52 <I440r> but i put them in anyway heh
23:43:11 <kholmes> hmm
23:43:15 <Xuz> I like static checks for some things.  And you at least need to make that your program isn't going to crash, no matter what the user types in.
23:43:17 <I440r> create and does> is the most powerful thing i have ever seen in any language
23:43:27 <I440r> it took me ages to get how does> works sorted out in my head heh
23:43:33 * Xuz notes that if you ask 5 forth programmers about something you'll get 9 answers.
23:43:37 <I440r> correct
23:43:56 <kholmes> is there anyway of assuring crash-proof programs?
23:44:03 <I440r> but teh kernel should allow the programmer to write ANY damned code he wants
23:44:05 <I440r> likle
23:44:10 <I440r> : foo if ......... ;
23:44:11 <kholmes> such as scheme and haskell can?
23:44:15 <I440r> : bar .......... then ;
23:44:36 * kholmes compiles gforth
23:44:40 <Xuz> If you want to write it.  THe program will let you access any chunk of memory you like.
23:44:44 <Xuz> Er, the system.
23:44:53 <Xuz> But, you can put any safeguards you like in there.
23:45:09 <I440r> so if you write bad code you are going to get sme sigsegv's heh
23:45:25 <Xuz> You could fiddle about with ! and @ and stick any checks you wanted in there.
23:45:56 <Xuz> And I've heard that some people were looking into getting a Forth system working with the Boehm GC, which would be. . . odd.
23:46:07 <I440r> which would only slow you down even more
23:46:14 <I440r> i rarely use @ and ! on variables
23:46:14 <aaronl> yeah
23:46:22 <I440r> i use constants and use ' >body ! heh
23:46:24 <I440r> faster
23:46:29 <kholmes> I440r: what do you use forth for?
23:46:48 <I440r> right now the only forth project i have is my own forth compiler for linux
23:46:50 <I440r> on hold
23:46:58 <I440r> cant get into it rite now :(
23:47:06 <I440r> got financial worries
23:47:32 <I440r> isforth
23:47:39 <I440r> want a linux x86 forth
23:47:45 <I440r> written entirely in assembler
23:47:51 <I440r> : definitions made with macos
23:47:55 <I440r> no libraries will be use
23:48:07 <I440r> everythign will be done with syscalls in the kernel itself
23:48:35 <Xuz> I'm more interested in somewhat the opposite, getting an ASM forth and then using that to write an OS kernel.
23:48:40 <Xuz> Er, converse.
23:49:05 <kholmes> what about type checking?
23:49:07 <I440r> heh
23:49:56 <Xuz> Ummm. . . 
23:49:57 <I440r> no such thing
23:49:58 <Xuz> Well. . . 
23:50:02 <Xuz> We don't have types.
23:50:06 <kholmes> ahh
23:50:06 <Xuz> Unless youw ant them.
23:50:13 <I440r> if you want to store an ascii string in a variable - go for it heh
23:50:26 <I440r> forth is untyped
23:50:31 <I440r> there memory and theres data
23:50:46 <Xuz> Oddly enough there's no end of object systems (if you want them) that've been written using XT's and CREATE/DOES>
23:51:34 <kholmes> so...to make forth high level you have to almost write a high level language in forth first?
23:51:34 <I440r> u mean OOP ?
23:51:45 <I440r> no
23:51:50 <I440r> forth IS a high level language :P
23:52:08 <kholmes> hmm...okay, what do you mean by high level?
23:52:25 <I440r> what do YOU mean ?
23:52:31 <Xuz> Look at it this way.  You know the relation between Scheme and Common Lisp?
23:52:34 <I440r> theres assembler = low
23:52:38 <kholmes> Xuz: yes
23:53:06 <kholmes> I440r: well...such as lisp, haskell, ML, Java...etc
23:53:31 <Xuz> Scheme tends more towards giving you everything you'll need to make any specialized construct you'd like, while CL tends more towards giving you that, AND a bunch of specialized constructs.
23:53:48 <kholmes> I440r: there is type checking, garbage collection, functions as first-level objects, recursion etc
23:54:02 <kholmes> Xuz: yeah
23:54:21 <Xuz> Forth is somewhat similar, providing alot of high-level functionality that you can specialize as much as you want, without giving yout he specifics, while also letting you fiddle with the hardware as mucha s you like.
23:54:37 <I440r> yes
23:54:46 <I440r> forth is that high :P
23:54:58 <Xuz> You could think of it as a high-level system for manipulating low-level objects.
23:54:59 <I440r> but you can do low level stuff with it
23:55:00 <I440r> because it has low level primatives
23:55:12 <Xuz> Which is why I'd like to see it used for an OS kernel.
23:55:36 <kholmes> interesting
23:55:41 <Xuz> Depending on your application it might be perfect, or it might also give you headaches.
23:55:44 * kholmes types 2 3 * .
23:56:12 <kholmes> no registers yet so it can't be assembly :)
23:56:48 <I440r> but you can code assembler within forth
23:57:19 <Xuz> It's modelled on the notion of a stack machine.
23:57:22 <kholmes> that sounds like a strange feature
23:57:26 <I440r> you can even interleave assembelr and forth in the same definition
23:57:26 <I440r> (;code) and (;uses) do this
23:57:34 <kholmes> reminds me of poking machine code from BASIC
23:57:38 <I440r> which are used by does> :)
23:57:41 <Xuz> Well, it's alot nicer.
23:58:03 <kholmes> whats the relationship between forth and ps?
23:58:23 <Xuz> And Forth does have all the features that you listed as high-level systems as having, except for garbage collection and type checking, and you would lose its ability to mess with low-level things if it did.
23:58:33 <Xuz> They're both stack-based.
23:58:50 <Xuz> PS is more specialized, and you don't get as much into hardware details.
23:59:06 <Xuz> Also if you redefine a word, all the words that use it will change their behaviour in PS, but not in Forth.
23:59:22 <kholmes> oh..hm
23:59:59 --- log: ended forth/02.01.16